close

Вход

Забыли?

вход по аккаунту

?

The immunologic detection and characterization of cartilage proteoglycan degradation products in synovial fluids of patients with arthritis.

код для вставкиСкачать
5 19
THE IMMUNOLOGIC DETECTION AND
CHARACTERIZATION OF CARTILAGE
PROTEOGLYCAN DEGRADATION PRODUCTS IN
SYNOVIAL FLUIDS OF PATIENTS WITH ARTHRITIS
JAMES WITTER, PETER J. ROUGHLEY, CAROLYN WEBBER, NANCY ROBERTS,
EDWARD KEYSTONE, and A. ROBIN POOLE
Antibodies were used in radioimmunoassayswith
gel chromatography to detect the hyaluronic acidbinding region, core protein, and keratan sulfate of
human cartilage protoeglycan in the synovial fluids of
patients with rheumatoid arthritis, juvenile rheumatoid
arthritis, and osteoarthritis. All fluids contained
proteoglycan that was mainly included on Sepharose
CL4B; this result indicates cleavage of proteoglycan
(which is normally excluded). The hyaluronic acidbinding region was the smallest and most commonly
detected fragment. It was relatively free of keratan
sulfate and core protein, and it could sometimes bind to
hyaluronic acid. Other larger fragments containing core
protein and/or keratan sulfate were detected in every
fluid.
From the Joint Diseases Laboratory, Shriners Hospital for
Crippled Children, the Department of Experimental Surgery, McGill
University, Montreal, Quebec, Canada, and the Rheumatic Disease
Unit, Wellesley Hospital, Toronto, Ontario, Canada.
Supported by the Arthritis Society of Canada and the
Shriners of North America. Dr. Witter was a Fellow of the Medical
Research Council of Canada, which also funded part of the work of
Drs. Poole and Roughley.
James Witter, PhD: Postdoctoral Fellow, Medical Research
Council of Canada, Joint Diseases Laboratory, Shriners Hospital for
Crippled Children (current address: Department of Surgery, Medical College of Wisconsin, Milwaukee); Peter J. Roughley, PhD:
Scientist and Associate Professor of Experimental Surgery, McGill
University, and recipient of a Chercheur Boursier award from
Fonds de la Recherche en Sante du Quebec; Carolyn Webber, MSc:
Graduate Student, McGill University; Nancy Roberts, BSc: Research Technician, McGill University; Edward Keystone, MD,
FRCP(C): Assistant Professor of Medicine, University of Toronto;
A. Robin Poole, PhD, DSc: Director, Joint Diseases Laboratory,
and Professor of Experimental Surgery, McGill University.
Address reprint requests to Dr. A. Robin Poole, Joint
Diseases Laboratory, Hospital for Crippled Children, 1529 Cedar
Avenue, Montreal, Quebec, Canada H3G 1A6.
Submitted for publication February 24, 1986; accepted in
revised form July 8, 1986.
Arthritis and Rheumatism, Vol. 30, No. 5 (May 1987)
All arthritides are characterized by the extensive destruction of articular cartilage, which leads to a
loss of joint function. Articular cartilage is primarily
composed of collagen and proteoglycan. Early in the
degradation process, proteoglycan is degraded and
released from cartilage (1,2), probably by proteinases,
which together, can cleave the core protein at various
sites (3). Cartilage proteoglycan degradation products
produced in vivo have not been well characterized.
Proteoglycans are composed of a protein core
to which glycosaminoglycan side chains are covalently
attached. Cartilage proteoglycans contain as many as
100 chondroitin-4-sulfate and chondroitin-6-sulfate
chains, with an average molecular weight of 20,000,
and 30-60 keratan sulfate (KS) chains, with molecular
weights of 4,000-8,000 (4). The overall molecular
weight of the largest proteoglycan is approximately
2-3 x lo6 (5). At one end of the cartilage core protein
is the hyaluronic acid-binding region (HABR), which
binds specifically to hyaluranic acid (HA) (6,7) to form
macromolecular aggregates that contain as many as
100 of these proteoglycan monomers (8).
The immunology of cartilage proteoglycans has
been recently studied in detail. Epitopes have been
identified on the HABR ( S l l ) , on other parts of the
core protein, including the chondroitin sulfate (CSF
and KS-rich regions (9), on CS oligosaccharides with
4,5-unsaturated glucuronosyl residues at the nonreducing end (12), and in association with KS (13-15).
Epitopes found on high buoyant density cartilage
proteoglycans are also present in the sclera and the
central nervous system (16). HABR is present in
arterial vessels (17), and KS is present in the cornea
(13,14,18). The cartilage proteoglycan epitopes associated with HABR, core protein, and KS are mainly
WITTER ET AL
520
Table 1.
Summary of the clinical histories of the 8 patients studied
Joint findings5
Patientdiagnosis*
Disease
SeroDisease duration
positive? Agehex activity$ (years)
C-RA
33lF
S
6.0
J-RA
SllF
MD
1.5
S-RA
63/M
MD
2.0
W-JRA (PL)
21lF
MD
0.75
B-JRA (P)
R-OA
2-OA
S-OA
17/M
24/F
55/F
81/M
M
MD
S
S
3.0
10.0
0.6
5.0
Drugs and dosage
Gold 50 mglweek,
Orudis 200 mgiday
Naprosyn 400-1,000 mg/
day
Entrophen 3,900 mgiday,
gold 10 mg/2 weeks
Naprosyn 1,000 mgiday,
penicillamine 250 mg/
day
Indomethacin 50 mg/day
Feldene 20 mg/day
Naprosyn 750 mglday
None
Joint
space
Erosions narrowing
+
+
ND
+
+
-
* RA
= rheumatoid arthritis; JRA = juvenile rheumatoid arthritis; PL = polyarticular onset; P =
pauciarticular onset; OA = osteoarthritis.
? - = no; + = yes; ND = not done; 2 = slightly reactive.
t S = severe; MD = moderate; M = mild.
; space
5 Determined by roentgenography. Erosions were either present (+) or absent (4joint
narrowing was either absent (-), moderate (++), or limited (+). ND = not determined.
concentrated in the cartilage matrix within the joint.
Thus, we are able to detect and characterize, by
immlunologic methods, cartilage proteoglycans and
their fragments in synovial fluid.
Earlier biochemical work revealed the presence
of chondroitin sulfate in synovial fluid (19-21) and
showed that there were increases in the CS content
relative to that of hyaluronic acid (2Q,21). Others have
described cartilage proteoglycan-like fragments in
normal ox synovial fluid of molecular weight 250,000
(22). Recently, using a polyclonal antibody to cartilage
proteoglycan, increased amounts of proteoglycan
were detected by enzyme-linked immunosorbent assay in the synovial fluids of patients with reactive
arthritis (23). No analyses were made to determine the
nature of the proteoglycan in the synovial fluids.
A careful study of cartilage proteoglycans in the
synovial fluids of patients with arthritis is of value for
two reasons. First, by identifying the types of fragments produced, we would gain some insight into the
mechanisms whereby proteoglycan is degraded. Second, a clear demonstration of cartilage proteoglycanderived fragments would provide us with a basis to
extend these studies to plasma and urine so that we
can attempt to develop new methods of detecting and
mopitoring cartilage degradation in disease and during
therapy. We have studied synovial fluids from 8 patients with rheumatoid arthritis (RA), juvenile rheumatoid arthritis (JRA), or osteoarthritis (OA), by use of
well-characterized antibodies to cartilage proteoglycans. The results of our study are reported here.
PATIENTS AND METHODS
Patient population. Patients were classified according
to American Pheumatism Association disease criteria
(24-26). Their clinical histories are described in Table 1 .
Because these patients all had relatively large knee effusions, they should be considered as a specific subgroup.
Synovial fluids. Synovial fluids were aspirated aseptically from patients’ knees into vacutainer tubes containing
EDTA (Becton Dickinson, Mississauga, Ontario, Canada).
The fluid was centrifuged at 343g for 10 minutes at room
temperature to remove cells and any particulate debris.
Fluids were stored frozen until used.
Hyaluronic acid was digested by incubation with
Streptomyces hyaluronidase (Calbiochem, San Diego, CA),
which can degrade HA only (27). To each volume of fluid
was added an equal volume of 100 mM sodium acetate
buffer, pH 5.0, containing 0.4 turbidity reducing units/ml of
Streptomyces hyaluronidase, together with proteinase inhibitors (as a precaution against proteoglycan degradation in
vitro), as was used for cartilage extraction described elsewhere (28). (Addition of 35S-biosynthetically labeled cartilage proteoglycan [a gift from Dr. L. Melching, Joint Diseases Laboratory] to synovial fluid and incubation at 37°C
for 24 hours revealed no degradation, as analyzed by gel
chromatography on Sepharose CL-2B.) Except for pepstatin, which was added at 10 pg/ml, each inhibitor was at a 2
mM concentration. Phenylmethylsulfonyl fluoride (PMSF;
used as a stock solution) was freshly prepared in isopropanol
(0.2M), which was diluted in the buffer. After 18 hours of
-i
CARTILAGE PROTEOGLYCAN DEGRADATION
incubation at 37"C, urea was added to a 6M concentration to
produce a clear yellow fluid.
Proteoglycan purification, biochemical analysis, and
treatment. High buoyant density proteoglycan from macroscopically healthy femoral or tibia1 adult human articular
cartilage was isolated from tissue temoved at autopsy, as
described previously (28,29). Proteoglycan monomer was
isolated either as an AlDl or a D1 preparation (28,29).
Pepsin and papain digestion fragments of AlDl and D1 were
prepared as described earlier (29). Where indicated, proteoglycan isolated as an AlDl preparation from human adult
articular cartilage was reduced and alkylated in 4M
guanidine hydrochlonde, using dithiothreitol followed by a
molar excess of iodoacetamide (13, and then dialyzed
against 100 mM Tris-acetate buffer, pH 7.3.
Human cartilage proteoglycan was treated with
proteinases as follows. To 2 mg of proteoglycan monomer in
850 pl of buffer was added 50 @I of enzyme solution: 27.5 pg
(13 units) of TPCK-trypsin (Millipore, Bedford, MA) in 50
mM Tris- HCl, pH 7.5; 10 pg of pepsin (Sigma, St. Louis,
MO) in 0.2M sodium acetate, pH 5.0; and 20 pg of papain
(Sigma) in 0.2M sodium acetate, pH 5.0, containing 5 mM
EDTA and 5 mM cysteine. After incubation for 24 hours at
3 7 T , inhibitors were added in 50-pl volumes. Trypsin was
inhibited by the addition of 20 pg of soybean trypsin inhibitor (type 1-S; Sigma), pepsin was inhibited by the addition
of pepstatin A (Sigma) dissolved in 2M Tris to a final
concentration of 10 pg/ml, and papain was inhibited by the
addition of iodoacetamide to 10 mM in the same buffer.
Controls were processed in the same manner, except that
proteinase was added at the end of the incubation with the
inhibitor.
Antibodies. Rabbit antiserum R114. Rabbit antiserum to native adult human proteoglycan isolated from
femoral articular cartilage was prepared as an AlDl preparation, as described above. Proteoglycan (4.5 mg) in 1.5 ml
of water was emulsified with 1.5 ml of Freund's complete
adjuvant (FCA; Difco, Detroit, MI) and injected intramuscularly, as 2 1.5-ml volumes, on day 0. The animal was
reinjected with half this amount of emulsion on days 20 and
36. On day 108,4 mg was injected in 1 ml of water emulsified
with 1 ml of Freund's incomplete adjuvant (FIA). On day
121, the animal was exsanguinated by cardiac puncture,
under anesthesia. The antiserum did not react with unsaturated oligosacchandes of HA and CS, detected as described
previously (30).
A typical radioimmunoassay (RIA) inhibition profile
using chondroitinase ABC-treated adult human cartilage
proteoglycan is shown in Figure 1. Antiserum R114 reacted
well with undigested (native) proteoglycan and with proteoglycan treated with chondroitinase ABC or keratanase.
Treatment of the proteoglycan with trypsin or pepsin considerably reduced the proteoglycan reactivity; papain totally
abolished its reactivity (Figure 1). The proteinase controls,
to which enzyme, followed immediately by inhibitor, was
added at the end of the incubation, gave the same inhibition
curve as those with no proteinase and no incubation. These
observations indicate that antiserum R114 reacts with an
epitope or epitopes that are wholly or partly protein in nature
or are protein-dependent. The fact that papain-cleaved
proteoglycan showed no reactivity also indicates that the
z8
52 1
:
,
/
40
or undigested
20
I
0.01
I
o.io
I
1-.o
I
5:O
PROTEOGLYCAN (pg)
Figure 1 . Radioimmunoassay of adult cartilage proteoglycan with
antiserum R114. Binding of antibody to 'ZSI-labeledchondroitinase
ABC (Chase ABC)+iigested proteoglycan was inhibited to various
degrees by unlabeled proteoglycan after treatment with the enzymes
indicated. Values are pg of proteoglycan per total assay volume.
antiserum does not react with free CS chaids or KS chains.
Thus, this antiserum is subsequently described as being
directed against the core protein of cartilage proteoglycan.
Rabbit antiserum R103. R103 antiserum represents
pooled sera from 3 rabbits immunized with purified HABR (a
generous gift from Drs. V. C. Hascall and J. Kimura,
National Institute of Dental Research, NIH, Bethesda, MD).
HABR was isolated from the Swarm rat chondrosarcoma
proteoglycan, as described previously (3 l), with clostripain
digestion of the aggregate to produce a complex of HABR,
link protein, and hyaluronic acid. HABR was separated from
link protein using a method previously described (31). On
day 0, each rabbit was injected intramuscularly with 0.37 mg
of protein in 1 ml of water emulsified with 1 ml of FCA. On
day 32, the injections were repeated. On day 117,0.20 mg of
protein emulsified in 2 ml of water with 3 ml of FIA was
injected intramuscularly. The animals were exsanguinated
by cardiac puncture on day 130.
A typical RIA inhibition profile using adult human
cartilage proteoglycan is shown in Figure 2. The reaction
with cartilage proteoglycan was considerably lower when
the proteoglycan was reduced and alkylated. Since the
HABR is the only part of the molecule where disulfide
bridges have been found (6), this loss of immunoreactivity is
indicative of the reaction of some of the antibodies with the
native conformation of HABR. Others have described a
similar rabbit antiserum to HABk, the binding of which is
inhibited when proteoglycan is reduced and alkylated (32).
Monoclonal antibody (AN9PI) to keratan sulfate.
This antibody (IgG1) was prepared by immunization of a
mouse with native adult human proteoglycan, isolated as D1
as described above. The preparation and specificity of this
antibody are essentially as we have noted for another
monoclonal antibody to KS (KPC-190) (ref. 15 and Webber
C, Glant T, Roughley PJ, Poole AR: unpublished observations). The antibody shows no significant reaction with
isolated KS chains, but it does react when the chains are
bound to core protein. A distinguishing feature of this and
other antibodies to keratan sulfate is that their reaction with
WITTER ET AL
522
'0°7
60
20
I
0.001
I
0.01
I
0.10
PROTEOGLYCAN (bg)
7
1.o
Figure 2. Radioirnrnunoassay of adult cartilage proteoglycan with
proteoglycan
antiserum R103. Binding of antibody to 1Z51-labeled
was inhibited by unlabeled proteoglycan (0)and, to a lesser degree,
by proteoglycan that had been reduced and alkylated (0).Values
are pg of proteoglycan per total assay volume.
KS as part of the native proteoglycan monomer is significantly inhibited when the monomer is pretreated with
keratanase, which leads to the partial digestion of KS (15). A
typical RIA inhibition profile is shown in Figure 3 . Further
characterization of these antisera and antibodies is discussed
below, as it relates to the results obtained from their use in
this study.
Radioimmunoassays. Proteoglycan labeling. One
hundred micrograms of adult human proteoglycan was labeled with Na'251 by the lactoperoxidase/glucose oxidase
method using Enzymobeads (Bio-Rad, Mississauga, Ontario, Canada). To a 50-p1 suspension of beads was added 100
pg of proteoglycan in 75 pl of 0.2M potassium phosphate
buffer, pH 7.2. One millicurie of NaIz5I (New England
Nuclear, Lachine, Quebec, Canada) was added in 20 pl of
0.4M sodium phosphate buffer, pH 7.4. D-glucose (25 pl of a
1.0% solution in distilled water) was added to initiate the
reaction. After 25 minutes, the reaction was terminated by
chromatography on Sephadex G-25 (1 x 10 cm) in 150 mM
NaCI, 50 mM Tris, pH 7.5, with 0.5 mg/ml of bovine serum
albumin (BSA; RIA grade).
Labeled proteoglycan was collected as the first
eluting peak. It was treated in the same solution, with
chondroitinase ABC at 0.025 unitshl, for 4.5 hours at 37°C
in the presence of the proteinase inhibitors, 1 mM iodoacetamide, 1 mM PMSF, 1 mM EDTA, and 5 pg/ml of pepstatin.
The sample was dialyzed overnight at 4"C, in dialysis tubing
with a 12,000-molecular weight cut-off (Spectrapor, Los
Angehs, CA), against 150 mM NaCI, 50 mM Tris buffer, pH
7.5, containing 0.05% NaN3. Samples were stored at 4°C
until used.
Assay. The incubation mixture contained 10 pl of
'"I-labeled proteoglycan (approximately 10,000 counts per
minute) in 150 pl of phosphate buffered saline (lo), with 2
mglml of BSA (RIA buffer), 25 pl of standard proteoglycan
or unknown sample (both were chondroitinase ABC-treated
before assay, by incubation with 0.025 units of chondroitinase ABC per ml in 50 mM Tris-acetate buffer, pH 7.3, at 37°C
for 3 hours with the above-mentioned protease inhibitors),
and 25 pl of antibody (or nonimmune immunoglobulin)
sufficient to bind approximately 30% of the radiolabeled
proteoglycan.
The mixture was incubated for 1 hour at 37"C, with
agitation. Fifty microliters of a 10% suspension of protein A
(Zysorbin, fixed and killed Staphylococcus aureus bearing
protein A; Zymed, Burlingame, CA) in 10 mM phosphate
buffered saline, pH 7.2, was then added. After 30 minutes at
37"C, the mixture was centrifuged at 1,500g for 10 minutes at
4°C. The supernatant was removed, and the pellet was
resuspended in 150 pl of RIA buffer and recentrifuged. The
supernatant was removed and the pellet was counted on a
gamma counter. Assays were performed in triplicate.
Binding curves were determined first by varying the
amount of antibody used in the absence of unlabeled
proteoglycan standard or unknown samples. Inhibition assays were performed with sufficient antibody to provide 50%
of maximal binding of the radiolabeled proteoglycan. Immune binding was determined by subtracting the binding
attained with the same amount of nonimmune control immunoglobulin. The percent inhibition of binding was calculated relative to the amount of '"I-labeled proteoglycan that
was bound in the presence of added inhibitor. Standard
inhibition curves were constructed for each assay to permit
conversion of the percent inhibition values to equivalents of
native (intact) proteoglycan. Typical standard inhibition
curves for the antibodies used are shown in Figures 1-3.
Ordinarily, chondroitinase ABC-treated adult human
proteoglycan was used to construct standard inhibition
curves. The specificities of these radioimmunoassays are
defined by the nature of the radiolabeled antigen, which in
this case, is purified cartilage proteoglycan monomer.
For RIA, monoclonal antibody AN9PI was used
after a single precipitation with 50% ammonium sulfate, as
described (33): antisera were diluted. Controls were nonimmune rabbit serum at the same serum dilution, or ammonium
sulfate-concentrated serum or nonimmune ascites fluid
(Bethesda Research, Gaithersburg, MD) at the same immunoglobulin concentration, free of naturally occumng antibodies to the nonreducing 4,5-unsaturated residues of
oligosaccharides of HA and C s , which we described recently (30). hnunoglobulin concentrations in ammonium
c/
s
.
,
,
.
'
I
40
20
0.001
0.01
0.10
1.o
PROTEOGLYCAN (pg)
Figure 3. Radioirnmunoassay of adult cartilage proteoglycan with
antibody AN9PI. Binding of antibody to 1251-labeledproteoglycan
was inhibited by unlabeled proteoglycan. Values are pg of proteoglycan per total assay volume.
523
CARTILAGE PROTEOGLYCAN DEGRADATION
sulfate concentrates were determined spectrophotometrically, as described (33). All assays, including those of column
fractions, were performed in triplicate. Means were recorded. The individual assay variation was 5-10%.
Ion exchange chromatography. DEAE-cellulose
(DE-52; Whatman, Hillsboro, OR) was equilibrated with 6M
urea in 50 mM sodium acetate buffer, pH 5.8. Ten milliliters
of synovial fluid was added per 1 gm of resin in a 1.5-cm
(diameter) column, which was chromatographed at 50
mYhour at room temperature. The column was washed with
2 bed volumes of 6M urea in 50 mM acetate buffer. It was
then eluted with 2 bed volumes of this solution containing
0.1M increments of NaCl from 0.1M to f.OM, followed by
elution with urealbuffer containing 1.5M and 2.OM NaC1. A
total of 26 fractions were collected, each containing 1 bed
volume. Fraction volumes of 1 ml were dialyzed against
deionized water using 3,500 molecular weight cutoff dialysis
tubing (Spectrapor). For RIA, 500-pl volumes of fractions
were incubated for 3 hours at 36°C with an equal volume of
200 mM Tris-sodium acetate buffer, pH 7.3, containing
chondroitinase ABC (Miles, Elkhart, IN), at 0.02 unitdml,
with proteinase inhibitors as described above. Because
immunoreactive material was present in all eluates, these
were pooled, dialyzed against deionized water as before, and
concentrated by lyophilization.
Gel chromatography. Twenty-five milligrams of the
material from ion exchange chromatography was rehydrated
for 24 hours at 4°C in 1.0 ml of 50 mM Tris-HC1 buffer, pH
7.5, containing 0.5M NaCl. This was chromatographed at
room temperature on Sepharose CL-4B (1.0 cm x 113 cm) at
a flow rate of 6.0 ml/hour; 1.0-ml fractions were collected.
For RIA, column fractions were treated with chondroitinase
ABC, as described above. The column was calibrated with
purified adult human proteoglycan (Dl), papain and pepsin
fragments of this proteoglycan (prepared as described
above), and BSA.
Aggregation of HABR. Two patients with RA (C-RA
and S-RA) were examined to determine whether the
proteoglycan fragments detected on Sepharose CL-4B contained functional HABR, which could bind to HA. Fractions
that eluted from Sepharose CL-4B between K,, 0 and K,,
0.6 were pooled, dialyzed at 4°C against deionized water for
18 hours, and lyophilized. The samples (4.2 mg and 9.9 mg
for C-RA and S-RA, respectively) contained total intact
proteoglycan equivalents of 71.2 pg and 330 pg for C-RA and
S-RA, respectively, as determined by RIA using antiserum
R103. They were dissolved in 2 ml of 4M guanidinium
chloride with 100 mM sodium acetate, pH 6.0, to dissociate
any HABR from HA oligosaccharides.
After 18 hours at 4"C, each of the samples was
divided into 2 equal volumes. To one volume was added HA,
in 16 pl or 35 pl of deionized water, to final concentrations of
2% (C-RA) and 10% (S-RA). To the other volume (controls)
was added deionized water only. After dialysis for 18 hours
at 4°C against 50 mM Tris-O.5M NaCl, pH 7.5, the solutions
were rechromatographed on Sepharose CL-4B in the same
buffer. Fractions were assayed using the RIAs for HABR
(antiserum R103), KS (monoclonal antibody AN9P1), and
core protein (antiserum R114), after treatment with
chondroitinase ABC, as described above.
Table 2. Summary of synovial fluids studied and their proteoglycan contents, determined by radioimmunoassay*
Synovial Yield
from
fluid
Patientdiagnosis
40
30
60
45
45
18
30
30
C-RA
J-RA
S-RA
W-JRA
B-JRA
R-OA
Z-OA
S-OA
~
volume
(ml)
Native proteoglycan equivalent
(dml)
DE-52 Monoclonal Antiserum Antiserum
R103
R114
AN9PI
(mg)
26.3
6.3
4.8
88
17.8
5.7
4.2
95
47.6
46.3
38.6
96
30.8
34.5
20.9
69
10.2
3.5
5.1
114
35.8
16.2
54
19.0
24.2
121.0
48.4
57
33.7
2.6
3 .O
45
~~
* Proteoglycan fragments eluted from DEAE-cellulose (DE-52)
were pooled and assayed to determine native proteoglycan equivalents (see Patients and Methods for details). RA = rheumatoid
arthritis; JRA = juvenile rheumatoid arthritis; OA = osteoarthritis.
RESULTS
Relative contents of proteoglycan fragments in
synovial fluids. Table 2 shows the patients studied, the
volume of synovial fluid obtained from each, and the
total yields (dry weight) of material bound to the ion
exchange column. Dry weight contents varied between 1.5 mg/ml and 3.0 mg/ml of synovial fluid. These
samples were assayed with antibodies AN9P1, R114,
and R103, which are directed against KS, proteoglycan core protein, and H A B R , respectively. Results
were expressed in equivalents of intact native
proteoglycan to permit an assessment of how much
equivalent native proteoglycan was present in these
fluids and to permit recognition of whether there was
any selective retention of a part of the molecule
released into the synovial fluid. This would be revealed by a higher native proteoglycan value compared with that obtained with other antibodies for a
given fluid.
The native proteoglycan equivalent values varied considerably, from 2.6 pg/ml to 121 pg/ml. The
results also show that in 6 of the 8 patients studied,
there was a greater proportion of the HABR compared
with the KS and core protein contents. Two of the
patients, 1 with RA (S-RA) and the other with JRA
(W-JRA), showed similar values for all 3 antibodies,
which suggests that those parts of the molecule that
are recognized by these antibodies are equally preserved. In general, the core protein and KS-rich
fragments were present in similar amounts in all patients.
Sepharose CL-4B gel chromatography findings.
Material bound to and eluted from DEAE-cellulose
524
WITTER ET AL
PG
BSA
PG
PG
AN9P1
lii_j”i_
10
.
0.0 0.2
0.8
0.4
0.6
0.8
J-RA
1.0
A114
0.0
U
o.o+
4
v
’
.
-
-
-
‘
I
AN9P1
:::
L
0.2 0.4
0.6
0.8
1.0
W
8
4
ANBPl
A
4
2
0
ANBPl
0.0
AlC
l3
10
0.0
0.2 0.4
0.6
S-RA
12
0.8
I
I
I
1.0
0.0
0.2
0.4
0.6
0.8
1.0
A114
4
0.4
0.6
0.8
1.0
0.0
10
0
0.0
KaV
0.0 0.2
0.1
0
A
!k
20
0.2
0.4
,K
B
0.6
0.8
1.0
;L
0.0
0.2 0.4
0.6
0.8
1.0
Kav
c
Figure 4. Gel chromatography of synovial fluids from patients with A, rheumatoid arthritis (patients C-RA,
J-RA, and S-RA); B, juvenile rheumatoid arthritis (patients W-JRA and B-JRA); or C, osteoarthritis (patients
R-OA, 2-OA, and S-OA). After binding to DEAE-cellulose (DE-52), samples were concentrated and
chromatographed on Sepharose CL-4B. Column fractions were examined by radioimmunoassay using antibodies
R114, AN9P1, and R103. Results are expressed as pg of proteoglycan (PG) per ml of column fraction. The elution
positions of intact human adult PG, pepsin-derived fragments (Pepsin PG), bovine serum albumin (BSA), and
papain-derived fragments (Papain PG) are shown across the top.
CARTILAGE PROTEOGLYCAN DEGRADATION
was pooled, dialyzed against water, and lyophilized. It
was then rehydrated and chromatographed on
Sepharose CL-4B. Column fractions were assayed
with each antibody, and the results were again expressed as intact adult cartilage proteoglycan equivalents. The chromatograms are shown in Figures 4A-C,
and a comparative analysis for the 3 antibodies for
each patient is presented in Table 3. There were
apparent qualitative and quantitative differences between individual synovial fluids in terms of the sizes
and amounts of proteoglycans. Yet, some trends could
be identified. The elution positions of pepsin-derived
and papain-derived adult cartilage proteoglycan fragments, intact proteoglycan, and BSA are shown at the
tops of Figures 4A-C, for reference. The majority of
the immunoreactive material was included on Sepharose CL-4B, whereas intact proteoglycan was excluded. These fragments were usually of a size that
was intermediate between those produced by pepsin
and papain cleavage.
In general, the larger fragments contained core
protein and KS epitopes. Those of intermediate size
generally contained both KS and core protein. The
HABR epitope(s) was mainly present on species of the
smallest size, which were somewhat larger than BSA
(60,000 M J and similar in size to HABR from rat
chondrosarcoma (67,000 M,) (31) and from bovine
nasal cartilage (90,000 M,) (7). These fragments usually contained relatively little KS and core protein;
however, in the samples from OA patients, the HABR
was larger and contained a greater proportion of KS
than core protein. In the 3 RA patients (Figure 4A) and
the 2 JRA patients (Figure 4B), the profiles for HABR
were generally well separated from the profiles of the
larger fragments, namely, R114 and AN9P1. HABR
was relatively absent from the largest fragments, except in 1 OA patient (S-OA) (Figure 4C).
The profiles of R114 and AN9P1 were very
similar for the 2 JRA patients (Figure 4B) and for 2 of
the RA patients (S-RA and J-RA) (Figure 4A), which
indicates that these fragments contained both core
protein and KS. Yet, differences in the relative distribution of these epitopes in fragments from other
patients indicated that KS is not always codistributed
on the core protein with those regions recognized by
antiserum R114.
Aggregation of proteoglycan fragments with HA.
Sepharose CL-4B coIumn fractions of C-RA and S-RA
between K,, 0 and K,, 0.6 were pooled, lyophilized,
dissolved in 4M guanidine hydrochloride with and
without added HA, and dialyzed against Sepharose
525
CL-4B column buffer. When rechromatographed in
the same buffer and assayed for HABR using antibody
R103, the elution profile of C-RA did not change with
the addition of HA; thus, there was no evidence for the
presence of a functional HABR in C-RA, although
HABR epitopes were detectable. In contrast, S-RA
contained some HABR which bound to HA and eluted
at the void volume (Figure 5). Forty percent of the
total HABR detected was aggregated HABR. Moreover, the total amount of aggregated and nonaggregated HABR that was detected represented 360% more
than the total amount detected in the absence of HA.
This increase may be due in part to the increase in
immunoreactivity that occurs when HABR binds to
HA, as described earlier (34). This increased reactivity
might produce, in part, an overestimation of HABR
compared with its native unbound state. We are presently investigating this possibility. Analysis with antibodies AN9P1 and R114 indicated that KS and core
protein showed little or no aggregation with HA. This
result indicates that in this patient (S-RA), the functional HABR had been largely separated from the rest
of the molecule.
DISCUSSION
By using antibodies to epitopes on cartilage
proteoglycans that aggregate with HA, we have been
able to demonstrate that fragments of these molecules
are present in synovial fluids of patients with different
types of arthritis. As yet, we have been unable, for
ethical reasons, to obtain synovial fluids from healthy
people for comparisons of fragmentation patterns.
When cartilage proteoglycans are experimentally degraded with purified proteinases, a variety of
cleavages can occur on the core protein, depending
upon the proteinase (7,3 1,35-37). Papain digestion of
bovine cartilage proteoglycan produces single chondroitin sulfate chains that are attached to a small
peptide fragment, whereas pepsin produces larger
fragments containing about 10 chondroitin sulfate
chains that are attached to a segment of the core
protein (38,39). The fragments we observed were
usually of intermediate size between those produced
when the proteinases papain and pepsin were used.
One of the fragments most commonly found in
synovial fluid was the HABR, which was identified by
its immunoreactivity, size, and ability to sometimes
bind to HA. The apparent loss of HABR function,
which is sometimes observed, indicates that the
HABR could be damaged either in the cartilage or in
WITTER ET AL
Table 3. Summary of gel chromatography of synovial fluids*
Native proteoglycan equivalents
(dml)
Patient-diagnosis
K,,
C-RA
0.05
0.35
0.60
0.05
0.35
0.60
0.05
0.35
0.60
0.05
0.35
0.60
0.05
0.35
0.60
0.05
0.35
0.60
0.05
0.35
0.60
0.05
0.35
0.60
J-RA
S-RA
W-J RA
B-JRA
R-OA
Z-OA
S-OA
Monoclonal
AN9PI
0.8
I .3
0.5
0. I
1.7
0.2
6.8
9.4
0.7
1.5
3.0
0.3
0.5
17.8
0.5
5.0
14.0
3.6
8.0
26.5
6.0
0.8
5 .O
2.2
Antiserum
RI I4
1.7
0.9
0.2
0.3
0.6
0.1
7. I
6.3
1.O
18.0
24.0
4.0
0.6
8.0
0.3
2.7
0.8
0.2
11.5
7.3
0.7
I .2
I .3
0.5
Antiserum
R103
degradation of adult human articular cartilage, induced
by interleukin- 1, have revealed that larger fragments
are, in fact, produced than have been observed in
synovial fluid (Poole AR, Campbell I, Roberts N,
3
2
1
0
I .8
4.4
23.0
0.2
0.7
24.0
3.2
5.0
16.8
0.3
2.6
9.0
R103 +HA
0
12
0.6
9.6
27.5
2.7
13.0
14.0
11.0
40.0
71.0
5.0
22.0
12.2
* Proteoglycan fragments eluted from DEAE-cellulose (DE-52)
were pooled, concentrated, and chromatographed on Sepharose
CL-4B (see Patients and Methods for details). RA = rheumatoid
arthritis; JRA = juvenile rheumatoid arthritis; OA = osteoarthritis.
the synovial fluid, and yet be recognized by antibody.
That it could still be recognized by antibody in its
denatured state was indicated by the only partial
decrease in immunoreactivity when the HABR was
denatured and inactivated by reduction and alkylation,
as shown in Figure 2.
When proteoglycan is present in aggregate
form, attached to HA and stabilized by link protein,
the HABR is protected from proteolytic digestion
(7,31). The size of the HABR can vary from M , 65,000
to M , 125,000, depending on the proteinase and the
degree of proteolysis. Thus, the larger fragments of
HABK, which co-chromatographed with KS and core
protein, probably contain these additional parts of the
adjacent proteoglycan molecule. The increased size is
probably also the result of more limited proteolysis.
Proteinases from polymorphonuclear leukocytes, such
as elastase and cathepsin G (35,39), may produce
further digestion of fragments released from cartilage
by the activity of metalloproteinases released from
chondrocytes (36,37,4043). The latter are known to
produce much larger fragments, which cannot always
aggregate: the cleavage site in this case appears to be
adjacent to the HABR (44). Our studies of the in vitro
R103 -HA
8
4
0
52
16
\
12
R 1 1 4 +HA
0,
1
8
U
a
a.
4
0
0
0.25 0-50 0-75 1-00
Kav
Figure 5. The aggregation of the hyaluronic acid-binding region with
hyaluronic acid (HA) in a patient with rheumatoid arthritis (patient
S-RA). Samples were chromatographed on Sepharose CL-IB with (+)
or without (-) added HA. Column fractions were assayed, and the
results are expressed as jg of proteoglycan (F'G) per ml of column
fraction.
527
CARTILAGE PROTEOGLYCAN DEGRADATION
Golds E: unpublished observations). This would be
consistent with a secondary phase of digestion into
synovial fluid.
It is believed that the aggregating proteoglycans
normally are bound to hyaluronic acid. The observation that HABR is released into synovial fluid raises
the obvious question as to how it is released from the
HA. Since the HABR in synovial fluid is often unable
to bind to HA, this release could occur as a result of
damage to the HABR by excessive activity, be it
proteolytic or free radical damage; however, this has
not been definitely established. It remains to be elucidated how functionally active HABR is released, but
the fact that it is released raises the question of
whether there is a hyaluronidase that could cleave
HA. Such an enzyme operating at neutral pH has
never been convincingly demonstrated in cartilage
matrix.
The frequent association of core protein with
KS (but not with HABR) on Sepharose CL4B indicates that antiserum R114 recognizes regions of the
core protein which contain KS. In addition, antiserum
R114 recognizes regions in which KS detected by
AN9PI is relatively absent (and vice versa), which
indicates that KS and the core protein epitopes are not
always present in the same places on the core protein,
and that these antibodies recognize epitopes which are
structurally unrelated to each other and to HABR.
Different fragments could have been produced
by a number of proteinases known to occur in inflamed
joints (namely, the proteinases of chondrocytes and
synovial cells), such as cathepsin D ( 4 9 , cathepsin B
(46), and the proteoglycan-degrading metalloproteinases (42), and as indicated above, by proteinases from
polymorphonuclear leukocytes. Only further analysis
of these fragments will help us identify possible candidates, by more in-depth comparisons of these degradation products with fragments produced by these
purified proteinases. We must bear in mind that some
fragments may be further digested or rapidly endocytosed and, hence, not be detectable in synovial fluid;
also, our analyses reflect the products of the net
balance between degradation, release, and removal of
cartilage proteoglycans.
It is clear that the production of these cartilage
proteoglycan fragments could be related to disease
activity and to the type of arthritis, as reported previously (23). The detection of these cartilage proteoglycan epitopes in the blood of healthy subjects and in
subjects with disease (47) may enable us to monitor
cartilage erosion in those with arthritis and to detect
and assess not only disease activity, but also the
effects of therapy. Our current studies of animals and
humans, which will be reported in the future, indicate
that this may sometimes be possible.
ACKNOWLEDGMENTS
We thank Drs. J. Esdaile, C. Watts, H. Tannenbaum
(Montreal General Hospital), and M. Baron (Jewish General
Hospital, Montreal) for supplying the synovial fluids; Freda
Rowbotham for typing the manuscript; and Mark Lepik for
preparing the illustrations.
REFERENCES
1 Dingle JT, Horsfield P, Fell HB, Barratt MEJ: The
breakdown of proteoglycan and collagen induced in pig
articular cartilage in organ culture. Ann Rheum Dis
34:303-311, 1975
2. Hamerman D: Cartilage changes in the rheumatoid joint.
Clin Orthop 64:91-97, 1969
3. Barrett AJ: The enzymic degradation of cartilage matrix,
Dynamics of Connective Tissue Macromolecules. Edited by PMC Burleigh, AR Poole. Amsterdam, North
Holland, 1975, pp 189-215
4. Hascall VC: Interaction of cartilage proteoglycans with
hyaluronic acid. J Supramol Structure 7: 101-120, 1979
5. Hascall VC, Sajdera SW: Physical properties and polydispersity of proteoglycan from bovine nasal cartilage. J
Biol Chem 245:4920-4930, 1970
6. Hardingham TE, Muir IHM: The specific interaction of
hyaluronic acid with cartilage proteoglycans. Biochim
Biophys Acta 279:401405, 1972
7. Heinegsrd D, Hascall VC: Aggregation of cartilage
proteoglycans. 111. Characteristics of the proteins isolated from trypsin digests of aggregates. J Biol Chem
249:4250-4256, 1974
8. Rosenberg L, Hellman W, Kkeinschmidt AK: Electron
microscopic studies of proteoglycan aggregate from bovine articular cartilages. J Biol Chem 250: 1877-1883,
1975
9. Wieslander J, Heinegikd D: Immunochemical analysis
of cartilage proteoglycans: antigenic determinants of
substructures. Biochem J 179:35-45, 1979
10. Poole AR, Reiner A, Tang L-H, Rosenberg L: Proteoglycans from bovine nasal cartilage: immunochemical
studies of link protein. J Biol Chem 255:9295-9305, 1980
11. Kimura JH, Thonar EJ-M, Hascall VC, Reiner A, Poole
AR: Identification of core protein, an intermediate in
proteoglycan biosynthesis in cultured chondrocytes
from the Swarm rat chondrosarcoma. J Biol Chem
256:789@7897, 1981
12. Christner JE, Caterson B, Baker JR: Immunological
determinants of proteoglycans: antibodies against unsaturated oligosaccharide products of chondroitinase ABCdigested cartilage proteoglycans. J Biol Chem 255:
7102-7105, 1980
528
13. Conrad GW, Ager-Johnson P, Woo M-L: Antibodies
against the predominant glycosaminoglycan of the mamimalian cornea, keratan sulfate-I. J Biol Chem 257:
464-471, 1982
14. Caterson B, Christner JE, Baker JR: Identification of a
monoclonal antibody that specifically recognizes corneal
and skeletal keratan sulfate: monoclonal antibodies to
cartilage proteoglycan. J Biol Chem 258:8848-8854,1983
15. Glant TT, Mikecz K, Poole AR: Monoclonal antibodies
to different protein-related epitopes of human articular
cartilage proteoglycans. Biochem J 234:31-41, 1986
16. F’oole AR, Pidoux I, Reiner A, Coster L, Hassell JR:
Mammalian eyes and associated tissues contain molecules that are immunologically related to cartilage
proteoglycan and link protein. J Cell Biol 93:910-920,
1982
17. Gardell S, Baker J , €aterson B, Heineggrd D, RodCn L:
Link protein and a hyaluronic acid-binding region as
components of aorta proteoglycans. Biochem Biophys
Res Commun 95: 1823-1831, 1980
18. Meyer K, Linker A, Davidson EA, Weissman B: The
mucopolysaccharides of bovine cornea. J Biol Chem
2Cm511-616, 1953
19. Barker SA, Hawkins CF, Zfewins M: Mucopolysaccharides in synovial fluid: detection of chondroitin sulphate. Ann Rheum Dis 25:209-213, 1966
20. Sweet MBE: An ultracentrifugal analysis of synovial
fluid. S Afr Med J 45:1205-1206, 1971
21. Seppala PO, Karkkaihen J, Lehtonen A, Makisara P:
Chondroitin sulphate in normal and rheumatoid synovial
fluid. Clin Chem Acta 36549-553, 1972
22. Silpananta P, Dunstone JR, Ogston AG: Fractionation
of a hyaluronic acid preparation in a density gradient:
the isolation and identification of a chondroitin sulphate.
Biochem J 104:404-409, 1967
23. S a m e T , Heinegiird D, Wollheim FA, Pettersson H:
Difference in cartilage proteoglycan level in synovial
fluid in early rheumatoid arthritis and reactive arthritis.
Lancet If:127-128, 1985
24. Ropes MW, Benrtett GA, Cobb S, Jacox R, Jessar RA:
Revision of diagnostic criteria for rheumatoid arthritis.
Ann Rheum Dis 18:49-53, 1959
25. Brewer EJ Jr, Bass JC, Cassidy JT, Duran BS, Fink
CW, Jacobs JC, Markowitz M, Reynolds WE, Schaller
J, Stillman JS, Wallace SL: Criteria for the classification
of juvenile rheumatoid arthritis. Bull Rheum Dis 23:
712-719, 1972
26. Altman R, Asch E, Bloch D, Bole G, Borenstein D,
Brandt K , Christy W, Cooke TD, Greenwald R,
Hochberg M, Howell D, Kaplan D, Koopman W,
Longley S 111, Mankin H, McShane DJ, Medsger T Jr,
Meenan R, Mikkelsen W, Moskowitz R, Murphy W,
Rothschild B, Segal M, Sokoloff L, Wolfe F: Development of criteria for the classification and reporting of
WITTER ET AL
osteoarthritis: classification of osteoarthritis of the knee.
Arthritis Rheum 29: 103!%1049, 1986
27. Ohya T, Kaneko Y: Novel hyaluronidase from Streptomyces. Biochim Biophys Acta 198:607-609, 1970
28. Roughley PJ, Poole AR, Mort JS: The heterogeneity of
link proteins isolated from human articular cartilage
proteoglycan aggregates. J Biol Chem 257: 11908-1 1914,
1982
29. Roughley PJ, White RJ: Age-related changes in the
structure of the proteoglycan subunits from human
articular cartilage. J Biol Chem 255:217-224, 1980
30. Poole AR, Reiner A, Roughley PJ, Champion B: Rabbit
antibodies to degraded and intact glycosaminoglycans
which are naturally occurring and present in arthritic
rabbits. J Biol Chem 260:6020-6025, 198.5
31. Caputo CB, MacCallum DK, Kimura JH, Schrode J,
Hascall VC: Characterization of fragments produced by
clostripain digestioh of proteoglycans from the Swarm
rat chondrosarcoma. Arch Biochem Biophys 204:220233, 1980
32. Ratcliffe A, Fryer PR, Hardingham TE: The distribution
of aggregating proteoglycans in articular cartilage: comparison of quantitative immunoelectron microscopy
with radioimmunoassay and biochemical analysis. J
Histochem Cytochem 32: 193-201, 1984
33. Tang L-H, Rosenberg L , Reiner A, Poole AR: Proteoglycans from bovine nasal cartilage: properties of a
soluble form of link protein. J Biol Chem 254:1052310531, 1979
34. Thonar EJ-M, Kimura JH, Hascall VC, Poole AR:
Enzyme-linked immunosorbent assay analyses of the
hyaluronate-binding region and the link protein of proteoglycan aggregate. J Biol Chem 257: 14173-14180,1982
35. Keiser H, Greenwald RA, Feinsteifi G, Janoff A: Degradation of cartilage proteoglycan by human leukocyte
granule-proteases-a model of joint injury. 11. Degradation of isolated cartilage proteoglycan. J Clin Invest
57:625-632, 1976
36. Galloway WA, Murphy G, Sandy JD, Gavrilovic J,
Cawston TE, Reynolds JJ: Purification and characterization of a rabbit bone metalloproteinase that degrades
proteoglycan and other connective-tissue components.
Biochem J 209:741-752, 1983
37. Morales TI, Kuettner KE: The properties of the neutral
proteinase released by primary chondrocyte culture and
its action on proteoglycan aggregate. giochim Biophys
Acta 705:92-101, 1982
38. Roughley PJ, Barrett AJ: The degradation of cartilage
proteoglycans by tissue proteinases: proteoglycan structure and its susceptibility to proteolysis. Biochem J
167:629-637, 1977
39. Roughley PJ: A comparative study of the glycosaminoglycan peptides obtained after degradation of
cartilage proteoglycan by different proteinases, and their
CARTILAGE PROTEOGLYCAN DEGRADATION
40.
41.
42.
43.
use in the characterization of different proteoglycans.
Connect Tissue Res 6: 145-153, 1978
Sapolsky AI, Malemud CJ, Norby DP, Moskowitz RM,
Matsuta K , Howell DS: Neutral proteinases from articular chondrocytes in culture. 2. Metal-dependent latent
neutral proteoglycanase and inhibitory activity. Biochim
Biophys Acta 658:138-147, 1981
Cartwright EC, Campbell IK, Britz ML, Sandy JD,
Lowther DA: Characterization of latent and active
forms of cartilage proteinases produced by normal immature rabbit articular cartilage in tissue culture. Arthritis Rheum 26:98&993, 1983
Woessner J F Jr, Selzer MG: Two latent metalloproteases of human articular cartilage that digest proteoglycan. J Biol Chem 259:3633-3638, 1984
Hauser P, Vaes G: Degradation of cartilage proteoglycans by a neutral proteinase secreted by rabbit bone-
529
marrow macrophages in culture. Biochem J 172:275284, 1978
44. Tyler JA: Chondrocyte-mediated depletion of articular
cartilage proteoglycans in vitro. Biochem J 225:493-507,
1985
45- Poole AR, Hembry RM, Dingle JT, Pinder I, Ring EFJ,
Cosh J: Secretion and localization of cathepsin D in
synovial tissues removed from rheumatoid and traumatized joints. Arthritis Rheum 19:1295-1307, 1976
46. Mort JS, Recklies AD, Poole AR: Extracellular presence of the lysosomal proteinase cathepsin B in rheumatoid synovium and its activity at neutral pH. Arthritis
Rheum 27509-515, 1984
47. Thonar EJ-MA, L~~~ME, K1intwoah GK, Caterson B,
Pachman LM, Glickman P, Katz R, Huff J, Kuettner
KE: Quantification of keratan sulfate in blood as a
marker of cartilage catabolism. Arthritis Rheum
28:1367-1376, 1985
Документ
Категория
Без категории
Просмотров
2
Размер файла
1 050 Кб
Теги
proteoglycans, degradation, patients, detection, arthritis, characterization, product, synovial, cartilage, immunologic, fluid
1/--страниц
Пожаловаться на содержимое документа